Choice of method for Seismic Analysis

[ThomasH]

Hi all

I have a couple of questions regarding the choice of metod for a seismic analysis in the US.

I have read some information and some sources suggest the lateral force method while others say dynamic approach and response spectra. One approach that I liked was that since the responce spectra method still has to refer to the lateral force metod for force levels (85% limit) you use the lateral force analysis based on several favorable assumptions. Then I do the design based om lateral force method and verify the assumptions with a response spectra analysis.

What I have is a large steel building. It's a factory with one or two storeys (it varies) and a large footprint. Typically the SFRS is Steel Special Moment Frames or Steel Special Concentrically Braced Frames. Would you use the lateral force method or the responce spectra method? Or a combination of these two?

Since the building is large we need to optimize it. My guess is that the simplifications in the lateral force method will result in a more conservative design, basically more material. Correct?

On a side note (pure curiosity ): If I want to run this nonlinear I would need a time-history accelerations for the site. Just out of curiosity (I very much doubt that I will need them), how can I find that type of data? I know there a differens methods to create "synthetic" data and I have seen a few options for software. But is there any common approach to do this? Or is the approach very rare?
I have worked with that type of data to create secondary spectra for very sensitive equipment. But in that case the client provided a site-specific spevtar with a given syntetic time-history.

Your input or comments would be appreciated.

Thomas

 

[MacGruber22]

My guess is that the simplifications in the lateral force method will result in a more conservative design, basically more material. Correct?

Probably not for the height of building you are talking about. 40-ft at most???

What is your footprint? We use Equiv. Lateral Force for large footprint (800 ft x 250 ft or so) 45-ft high tilt-up warehouses with moment or braced frames regularly.

I believe most of your economy will be to use a braced frame system in lieu of a moment frame system (if of course there aren't architectural restrictions), rather than your selection of analysis.

What is your Sds and Sd1?


"It is imperative Cunth doesn't get his hands on those codes."

 

[ThomasH]

MacGruber22

Thanks for your reply.

I'm have not been at the offece today but I can give you some data from memory.

The building layout is approximately 500 meters x 100 meters. Then we split the 100 meter width i 3 sections (30 + 40 + 30 meters). The section in the middle will have two floors (probably slightly higher 40 ft) while the two side sections will have one floor (probably less than 40 ft).

The reason for my question is that I read some of the technical briefs from NEHRP and I got the impression that response spectra analysis is frequently used:

http://www.nehrp.gov/library/techbriefs.htm

I also looked at a webinar that gave me the same impression:

https://www.aisc.org/UploadedContent/2008NASCCSessions/P5/default.htm

But that may be wrong. What, in your opinion, would be the advantages/disadvantages for response spectra vs lateral force method.

Best Regards

Thomas

 

[KootK]

X2 for MacGruber's comments. The main benefit of ELF is design effort efficiency. The main benefit of dynamic analysis is structural system optimization.

Consider that ELF basically reduces buildings to equivalent single degree of freedom (single story) structures. The closer your building is to actually being a single degree of freedom structure, the less benefit you're likely to get from a dynamic analysis. A one story building is a single degree of freedom system. And a two story building isn't too far off.

Consider the statements above to be general in nature. If you were to be exploring base isolation or supplemental damping, there may be advantages to using dynamic analysis, even with a single story structure.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[ThomasH]

KootK

I agree with your comments.

Lateral force is most likely "easier" to perform but dynamic analysis may result in a more optimized structure. I think that one way to check the possible effects of a dynamic analysis can be to check how much of the total mass is participating in the first mode.

I think that due to the size of the project some type of optimizing will be performed. But if the structure behaves as a single degree of freedom system there is not a lot of parameters to work with.
But I think it is probably better to use a period (flexural frequency) based on an analysis rather than the formulas in the code. On the other hand, if the formulas give a more advantageous result perhaps we should use them. That was one of the ideas from the mentioned wedinar.

We'll we what happens.

Anyway, Thanks for you input.

Thomas

 

[MacGruber22]

but dynamic analysis may result in a more optimized structure

Again, not so much for the structure you describe.

Look at it this way - I am betting a fee has already been provided and a contract signed for this project. If that fee doesn't provide for dynamic analysis, you will be working extra for nothing (other than for your own knowledge). Keeping in mind the man-hours, just run the ELF with the best possible configuration of a LFRS. If the owner/contractor want to optimize further once they get a hold of the pricing drawings, you have change order power to go back and revise. But, again, I think you are barking up the wrong tree with dynamic analysis. Well-positioned braced frames will be very economical verses moment frames.

"It is imperative Cunth doesn't get his hands on those codes."

 

[ThomasH]

Actually, I think that the fee would allow for a dynamic analysis. Provided that the dynamic analysis has a value for optimizing the final structure.

But I am not convinced that is has av value in this case. Since I work a lot with dynamic problems I wanted some input regarding how "common" that approach is for a seismic analysis in the US.

Thanks for your input

Thomas

 

[MacGruber22]

Actually, I think that the fee would allow for a dynamic analysis.

*think*? You should *know*. It sounds like you need to talk to the senior engineer(s) who developed the fee.

I wanted some input regarding how "common" that approach is for a seismic analysis in the US

The short answer is that you don't unless you have to

"It is imperative Cunth doesn't get his hands on those codes."

 

[ThomasH]

Sorry, bad choice of word on my behalf.

The fee allows for a dynamic analysis but before I/we start I wanted to get some ideas regarding if it would actually add anyting. (Or perhaps rather "remove" any material )

The project want us to optimize the structure. If that means that we use more engineering-hours, so be it. But we don't want to spend more money optimizing than we save in the steel (for example).

The reason I asked is that even though I have made several dynamic seismic analyses (both response spectra and a few time-history). But I have never really compared lateral force against response spectra for a fairly simple structure. Lateral force should be conservative but perhaps not so conservative that it pays to take things further.

Thanks

Thomas